University of British Columbia

Professor Jeff Clune's research lab at the University of British Columbia conducts work at the intersection of deep learning, open-ended AI systems, and AI safety. Clune joined UBC's Department of Computer Science as an Associate Professor in January 2021, following a role as research manager at OpenAI and prior work at Uber AI Labs and the University of Wyoming. His research program encompasses several broad areas. In machine learning, he is known for work on deep reinforcement learning, neuroevolution, and quality-diversity algorithms that produce large, diverse sets of high-quality solutions rather than optimizing a single objective. He is a primary developer of the AI-generating algorithms (AI-GAs) paradigm, which seeks to automate the process of creating AI systems by using AI itself to innovate and discover new architectures, learning algorithms, and training environments. On the AI safety side, Clune has expanded his research agenda to address alignment, interpretability (which he terms 'AI neuroscience'), and governance. In February 2023, he announced that his UBC lab received a grant from Open Philanthropy for work on AI alignment, safety, and existential risk. He advocates for prudent governance of advanced AI systems, including democratic coalitions, regulation of frontier models, and global alignment protocols. Clune also holds a Canada CIFAR AI Chair at the Vector Institute for Artificial Intelligence and serves as a Senior Research Advisor at DeepMind. His lab is affiliated with UBC's Centre for Artificial Intelligence Decision-making and Action (CAIDA). He has received numerous honors including the SIGEVO Impact Award (2023) and a Presidential Early Career Award from the United States (2019).